paper-with-me

Papers

Generalisation Guarantees for Continual Learning with Orthogonal Gradient Descent

2020-06-21 · Mehdi Abbana Bennani, Thang Doan, Masashi Sugiyama

In Continual Learning settings, deep neural networks are prone to Catastrophic Forgetting. Orthogonal Gradient Descent was proposed to tackle the challenge. However, no theoretical guarantees have been proven yet. We present a theoretical framework to study Continual Learning algorithms in the Neural Tangent Kernel regime. This framework comprises closed form expression of the model through tasks and proxies for Transfer Learning, generalisation and tasks similarity. In this framework, we prove that OGD is robust to Catastrophic Forgetting then derive the first generalisation bound for SGD and OGD for Continual Learning. Finally, we study the limits of this framework in practice for OGD and highlight the importance of the Neural Tangent Kernel variation for Continual Learning with OGD.

📄 PDF Abstract BibTeX arXiv:2006.11942

Code (1)

MehdiAbbanaBennani/continual-learning-ogdplus 공식 구현 pytorch

Tasks

Continual LearningTransfer Learning

Methods 이 논문이 사용한 방법론

SGD Stochastic Gradient Descent is an iterative optimization technique that uses minibatches of data to form an expectation of the gradient, rather than the full gradient using…

Similar Papers 제목 키워드 기반

ONG: Orthogonal Natural Gradient Descent

2025-08-24 · Yajat Yadav, Patrick Mendoza, Jathin Korrapati arxiv

Orthogonal Gradient Descent (OGD) has emerged as a powerful method for continual learning. However, its Euclidean projections do not leverage the underlying information-geometric structure of the problem, which can lead …

Continual LearningRotated MNIST

On the Theory of Continual Learning with Gradient Descent for Neural Networks

2025-10-07 · Hossein Taheri, Avishek Ghosh, Arya Mazumdar arxiv

Continual learning, the ability of a model to adapt to an ongoing sequence of tasks without forgetting earlier ones, is a central goal of artificial intelligence. To better understand its underlying mechanisms, we study …

Continual Learning

Fisher-Orthogonal Projected Natural Gradient Descent for Continual Learning

2026-01-19 · Ishir Garg, Neel Kolhe, Andy Peng, Rohan Gopalam arxiv

Continual learning aims to enable neural networks to acquire new knowledge on sequential tasks. However, the key challenge in such settings is to learn new tasks without catastrophically forgetting previously learned tas…

Continual Learning

Meta Mirror Descent: Optimiser Learning for Fast Convergence

2022-03-05 · Boyan Gao, Henry Gouk, Hae Beom Lee, Timothy M. Hospedales

Optimisers are an essential component for training machine learning models, and their design influences learning speed and generalisation. Several studies have attempted to learn more effective gradient-descent optimiser…

Meta-Learning

Debugging using Orthogonal Gradient Descent

2022-06-17 · Narsimha Chilkuri, Chris Eliasmith

In this report we consider the following problem: Given a trained model that is partially faulty, can we correct its behaviour without having to train the model from scratch? In other words, can we ``debug" neural networ…

Continual Learning